Me, too....one guy told me over drinks "The trick is to always hire guys smarter than you, and not to let that bother you...if they're NOT smarter than you, what's the point? The problems are getting them to all work together - it's like herding kittens on caffeine - and getting them to stay when they figure out what they're worth, or that they should start their own company."
Jim and Bob were the best. Used to run into Jim all the time at Apps seminars.
And it was always easy to tell where Bob was; his "Lightning Bug" was unmistakable. They will be missed.
I've recently readed in a people magazine in the dentist waiting room that Jean-Claude Van Damme was studiying physics in order to improve his understanding of the molecules.
I feel a glimmer of hope each time i open an interesting book, but i'm still don't know if i am intelligent or not.
I never had a thought of being stepped on... I am way too old for that. I worked at National with Jim in Linear IC, I worked on the Ford Ignition IC. I would like to see the schemo of that amp you referred to that beat the Leach. But let us all keep working and sharing this wonderful Hobby. I really appreciate this site. 😀
I've also heard that Anne Hathaway is a physics geek. Just watched "Interstellar" for the third time.
Her new husband is one lucky guy!
Last edited:
Way better than I thought it would be. Kip Thorne of LIGO was the technical advisor and insisted the physics fit current theory as much as possible. He and Dr. Weiss, and few other members of the team signed signed a copy of their Nobel paper as thanks for my (very) tiny contribution.
I figure I'm qualified now to sell quantum snake oil.
We used those overclocked to 12 blazing megahetz to run subtitles at film festivals. That was about 30 years ago. They were the only thing we had that could keep up. A good tool for the job at hand.
The movie is visually STUNNING...and an awesome soundtrack.
Feynman got the Nobel in '65 with Schwinger and Tomonaga. Thorne was Feynman Professor of Theoretical Physics at CalTech (where Feynman gave his now-famous Lectures Series), and got the prize for the LIGO team's announcement of the detection of gravitational waves.
I've read all of Feynman's stuff, but my personal favorites are the interview he gave on "The Pleasure of Finding Things Out" and his stark intellectual honesty - what I call the 'Feynman criterion' : if you can't explain something so that someone with a high-school education in math and physics can understand it, it's probably wrong (and he was talking about the state of modern quantum physics theories). At the time of his death, he was still frustrated by the inability of his breakthroughs in QED to be adapted to understanding the characteristics of the interactions occurring inside the nucleus. Quark Theory and Superstring Theory were BOTH proved to be inadequate by the results that came out of the Large Hadron Collider. One of the physicists involved joked that all they proved was that a bunch of really really smart people could convince the world to give then billions of dollars to build Earth's biggest elliptical magnet. Reminds me of high-end audio....
Member
Joined 2009
Paid Member
totally agree, that was a superb movie, worthy of having borrowed heavily from Kubrick's earlier work.
these days, the public have their insight into the character of the physics genius through Sheldon Cooper 😀
these days, the public have their insight into the character of the physics genius through Sheldon Cooper 😀
Just some goofy TV trivia. Do you know who the Big Bang characters Sheldon and Leonard were named for?
...30 years ago..1987...the Sun 3/60 on my desk at work was running a 20 MHz 68020...(32 bit)
....the PC on my desk at home was running a 33MHz Pentium III (16 bit).
I still have that one, and it still runs. I keep it around to interface to obsolete stuff like old ESDI hard disks and tape backups people need data retrieved from. It still has a copy of the original BattleChess game on it, and I think Ms PacMan. 8MB of main memory, and a whopping 500 MB SCSI hard drive.
I'm all for putting vintage computer hardware to good use. A 700 MHz Pentium III (yes, a '3' - and they were pushing 1.2 GHz when they were superseded by the P4, etc.) motherboard is all you need to make a box that will decode and play DVDs. And it's silent, because a simple, passive heatsink is enough to cool it. And it will run Windows 98 (don't laugh - that's what's running behind the touchscreens of some store 'Self-Checkout' systems - and because a running program can have complete control of the hardware, with no multitasking, it's FAST), or, for that matter, Windows XP. As far as my comment on still being 'unimpressed' goes...Apple products are EXTREMELY well-designed. It's just the cost/performance ratio compared to the PC is laughable. And Apple insists on making everything proprietary, and most often, not upgradeable. A few notable exceptions, the first major one being the Mac SE series, that offered ONE expansion slot. I worked at SuperMac Technology briefly, to ugrade the audio portion of their video & audio capture board that allowed the SE to be used for video editing, from 12bit(!) audio to a more modern 16 bit/44.1 KHz CD-standard resolution. Apple actually contracted Supermac to do other designs as well, because Supermac could get products designed and ready for production faster than Apple's own internal design teams. In '93 the guys spinning SRS Labs off of Hughes Aircraft recruited me, my replacement at Supermac couldn't complete the design in time, and Supermac got bought by Radius in '94.
I'm currently designing a new multiprocessor computer that will be optimized for streaming and processing audio and video in realtime, and will be able to run Linux or Windows, just so people can click 'Apps' to run it. It's based on my work to build something to run the tools from Stanford's CCRMA group designed for music synthesis and Acoustic WaveFront Imaging virtual environment processing. There's an interesting project somewhere on DIYAudio concerning using a PC to perform as a software-based configurable speaker Crossover/EQ box that's a good application for this, as well. No more hand-wound inductors or expensive precision foil capacitors. I think it does room correction as well.
What a concept! Do it ALL in software, and run the speaker elements directly, and individually.
Sorry for the long post - but this project and caffeine get my tail wagging every time.
As amazing as that is, we weren't about buy a Sun or Spark or whatever Unix workstation to do the task. We were already too expensive with our little Apple IIe machines.
And since we didn't have your time machine in 1986, we weren't able jump over to 1999 to buy a Pentium III. Heck, even the Pentium Pro was 9 years off and out of reach of our limited time machine. 😀
Some of us just have to work with what we have. The pain still lingers, tho.
High Street equipment has to built to a very strict budget, corners are cut to save pennies on each unit. DIY enables you to bring out the best in what you build.
I just love the learning experience too.
I just love the learning experience too.
Member
Joined 2009
Paid Member
It's in Analog Devices application note AN-211 "The Alexander Current-Feedback Audio Power Amplifier" available online at
http://www.analog.com/media/en/tech...tes/58052492001115525484056221917334AN211.pdf
or just Google "Mark Alexander current feedback audio amp" and it pops.
(the online scan is on ADI's own server and looks like it was scanned at low DPI - I get get you a better one if you need it.)
It's also covered by a patent, which really doesn't have much more info than the App Note, and the App Note has PCB layouts. I might have a couple of unstuffed boards laying around, if you're truly interested in building one. But you could do a high-quality scan from a better copy and Photoshop it to
produce masks good enough for exposing or direct inkjet printing PCBs. I do that for boards I do mods to - so much neater than jumper wires and parts hanging in mid-air.
Uses a single pair of IGBT's per channel, one N and one P, but they are mounted external to the input/driver PCB, which is less than 2.5 in. x 2.5 in. IIRC
Easily upgradeable to MOSFETs instead (I hate IGBTs) and more devices/rail for more power.
The whole patent thing is easily handled; all Analog Devices requires is that you use their parts (2 - an SSM-2131 and an OP-97, per channel), and that's only necessary if you want to sell them. Easily modded to use AD845, OPA627, AD797 - but I'd keep the OP-97 in there for simplicity's sake.
I played around with using a really expensive chopper-stabilized part instead, and didn't gain squat for the added expense. I also modded mine to replace a small signal-type reed relay in the bias circuit with an optoisolator, but that's just because mechanical parts like that are something I don't like to use,
except in IC test jigs. It is used for (and works quite well at) eliminating turn-on 'thump' - with NO !$%#! relays in the output. - My Klipsch La Scala speakers can testify to that. But my mods to my Hafler DH120 and ALL my preamps do that nicely as well with no tiny relay or opto parts. Just attention to what linear IC designers do to guarantee goof-free startup of Wilson current mirrors and folded-cascode translation stages in a primary gain node.
Mark's brilliant, and the app note covers EVERYTHING you need to know, including the math for changing the frequency response, tweaking the compensation for different output devices, as well as layout and wiring gotchas. You should have no problem following this stuff. You were around
when National was selling op-amps to Moses, and he was calling in to
Bob
or Jim 
for customer support, right?
Specs (per channel) - measured on an Audio Precision:
70 W out into 8 ohms (voltage limited)
0.001% (that's 10 parts-per-million) at 1 kKHz, 50 W, 8 ohms
rising to a whopping 0.009% at 20 KHz, 50w, 8 ohms
SMPTE Intermodulation distortion 0.0004% (4 ppm) at 41.7 W
-3dB DC to 1 MHz (you could use it as an AM transmitter, in a pinch)
slew rate >200v/uS
rise time 400 ns (yes - nano)
idle current 140-150 mA (This is possible because of the unique way
the precision OP-97 handles the DC behavior, while the SSM-2131
handles the AC behavior.)
DC output is drift-free with temperatue (adjustable, along with idle current)
Note - I have NO commercial interest in this, or stock in Analog Devices.
Just a lot of stock OF Analog Devices' parts, but that goes for National,
Burr-Brown (bought by TI), Linear Technology (now bought by Analog Devices) parts as well.
Sorry - should have said 33 MHz 386. And the Pentium Pro may have been 9 years off, but we were designing a prototype that we demonstrated at COMDEX (in 1990) that used a 66 MHz 486 tricked out to run a 66 MHz, 64-bit motherboard with quad-pumped interleaved 64bit memory (technically DDR2 performance, 9 years before RamBus, 10 years before even DDR). Which is why IBM gave us a check for $25 mil and AMD ended up buying us in '96, to get the tech that became the K6.
"You like apples....how you like THEM apples!" - 'Good Will Hunting'
- Status
- Not open for further replies.